Correlation Analysis Model of Social Capital and Innovation Performance Based on Knowledge Mapping

Area of Science:

• Business and Management
• Information Science
• Innovation Studies

Background:

• Social capital is crucial for Small and Medium-sized Enterprises' (SMEs) innovation performance.
• Existing recommendation models often underutilize the rich correlations within knowledge graphs.
• Understanding the mediating factors in the social capital-innovation performance link is vital for SMEs.

Purpose of the Study:

• To analyze the correlation between social capital and innovation performance using a knowledge graph approach.
• To propose a novel cross-attention fusion-based knowledge graph recommendation algorithm (CAKR).
• To investigate the mediating roles of knowledge management capabilities and environmental dynamics.

Main Methods:

• Development of a correlation analysis model leveraging knowledge graph technology.
• Implementation of the CAKR algorithm featuring cross-fusion, recommendation, and knowledge graph embedding modules.
• Empirical study examining the mediating effects of knowledge management capabilities and environmental dynamics on the social capital-innovation performance relationship.

Main Results:

• The CAKR model effectively utilizes knowledge graph correlations for enhanced recommendations.
• Knowledge management capabilities fully or partially mediate the relationship between different dimensions of social capital and innovation performance.
• Environmental dynamics positively mediate the link between knowledge management capability and innovation performance, with stronger dynamics enhancing the effect.

Conclusions:

• SMEs can strategically leverage social capital to boost innovation performance.
• Knowledge management capabilities and environmental dynamics are key factors influencing this relationship.
• The CAKR algorithm offers a promising approach for knowledge graph-based recommendations in innovation contexts.